Do you mean, that if I have a
class Bo {
IList<Thing> Things {get; set;}
}
and in client code I do something like
bo.Things = new List<Thing>()
At least that would not apply here - all my Collections are exposed as
ReadonlyCollection<T> and have AddThing(Thing t) etc.... methods.
Maybe I don't quite follow - could you explain with a bit more detail (or a
link to somewhere, where this is described?)
Thanks,
Jan
2009/7/2 Jason Meckley <[email protected]>
>
> Jan, I have seen other describe this as well. Usually issue is the
> collection is modified through the accesser. when NH checks for dirty
> entities it will save them. so while you don't want an update, NH
> believes one is required and saves the changes. this may be what is
> slowing the system down.
>
> for a better look at what is happening check out NH Profiler
> [www.nhprof.com]. This will provide a wealth of knowledge about what
> is happening.
>
> On Jul 2, 12:30 pm, Jan Limpens <[email protected]> wrote:
> > something weird as well...
> >
> > When I request a transaction decorated action, my log reads like this:
> > -- this repeats a lot of times! maybe for each returned row:
> >
> > 2009-07-02 13:18:39,665 [4] INFO NHibernate.Engine.Cascade - cascade
> > NHibernate.Engine.CascadingAction+SaveUpdateCascadingAction for
> collection:
> > Fex.Businesslayer.Theme.Articles
> > 2009-07-02 13:18:39,667 [4] INFO NHibernate.Engine.Cascade - done
> cascade
> > NHibernate.Engine.CascadingAction+SaveUpdateCascadingAction for
> collection:
> > Fex.Businesslayer.Theme.Articles
> > 2009-07-02 13:18:39,671 [4] INFO NHibernate.Engine.Cascade - done
> > processing cascade
> > NHibernate.Engine.CascadingAction+SaveUpdateCascadingAction for:
> > Fex.Businesslayer.Theme
> >
> > -- end loop
> >
> > 2009-07-02 13:18:39,689 [4] INFO Default - Request
> forhttp://localhost:15968/trial/somethingtd.asprtook 1054ms.
> >
> > -- this was a distributed transaction, it takes even more time
> >
> > -- now the following is ok!
> >
> > 2009-07-02 13:19:19,035 [9] INFO NHibernate.Loader.Loader - SELECT
> > this_.ThemeID as ThemeID11_0_, this_.BusinessKey as Business2_11_0_,
> > this_.Name as Name11_0_, this_.Description as Descript4_11_0_,
> this_.Visible
> > as Visible11_0_, this_.KitID as KitID11_0_, this_.AdditionalGuestKitID as
> > Addition7_11_0_, this_.AdditionalDecorationID as Addition8_11_0_,
> > this_.PartyFavorID as PartyFav9_11_0_, this_.ConversationID as
> > Convers10_11_0_ FROM Fex2.dbo.[Theme] this_ WHERE (@p0 < (SELECT
> > count(articles1_.ArticleID) as y0_ FROM Fex2.dbo.[Theme] this_0_ left
> outer
> > join Fex2.dbo.ArticleTheme articles3_ on
> this_0_.ThemeID=articles3_.ThemeID
> > left outer join Fex2.dbo.[Article] articles1_ on
> > articles3_.ArticleID=articles1_.ArticleID WHERE this_.ThemeID =
> > this_0_.ThemeID) and this_.Visible = @p1) ORDER BY this_.Name asc
> > 2009-07-02 13:19:19,098 [9] INFO Default - Request
> forhttp://localhost:15968/trial/something.asprtook 71ms.
> >
> > I find it strange this is logged at INFO (usually these are DEBUG), but
> it
> > might have something to do with the problem....
> >
> > It also could be a sideeffect of the unwanted logging :)
> >
> > 2009/7/2 Jan Limpens <[email protected]>
> >
> >
> >
> > > I don't actually know how sql server (in which format) receives the
> actual
> > > query.
> > > but padding the test query in Query Analyzer with 'begin transaction'
> and
> > > 'commit transaction' does not yield any differences in execution time.
> >
> > > 2009/7/2 Jan Limpens <[email protected]>
> >
> > > I could privately send you a dot.trace file. Can you open those?
> >
> > >> 2009/7/2 Tuna Toksoz <[email protected]>
> >
> > >> Do you have a profiler? or can you share this sample project with us
> so
> > >>> that we can profile?
> >
> > >>> Tuna Toksöz
> > >>> Eternal sunshine of the open source mind.
> >
> > >>>http://devlicio.us/blogs/tuna_toksoz
> > >>>http://tunatoksoz.com
> > >>>http://twitter.com/tehlike
> >
> > >>> On Thu, Jul 2, 2009 at 6:06 PM, Jan Limpens <[email protected]
> >wrote:
> >
> > >>>> 2009-07-02 12:02:52,362 [4] INFO Default - Request for
> > >>>>http://localhost:15968/trial/something.asprtook 125ms.
> > >>>> 2009-07-02 12:02:54,371 [4] INFO Default - Request for
> > >>>>http://localhost:15968/trial/something.asprtook 124ms.
> > >>>> 2009-07-02 12:02:55,115 [8] INFO Default - Request for
> > >>>>http://localhost:15968/trial/something.asprtook 116ms.
> > >>>> 2009-07-02 12:02:58,838 [8] INFO Default - Request for
> > >>>>http://localhost:15968/trial/somethingt.asprtook 477ms.
> > >>>> 2009-07-02 12:03:00,215 [4] INFO Default - Request for
> > >>>>http://localhost:15968/trial/somethingt.asprtook 406ms.
> > >>>> 2009-07-02 12:03:01,227 [8] INFO Default - Request for
> > >>>>http://localhost:15968/trial/somethingt.asprtook 411ms.
> >
> > >> --
> > >> Jan
> >
> > > --
> > > Jan
> >
> > --
> > Jan
> >
>
--
Jan
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Castle Project Users" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/castle-project-users?hl=en
-~----------~----~----~----~------~----~------~--~---
